A 0.12% parameter add-on gives AI agents the working memory RAG can't

Researchers built delta-mem to give AI agents working memory at 0.12% parameter overhead, outperforming RAG and context ...
VentureBeat

Google PM open-sources Always On Memory Agent, ditching vector databases for LLM-driven persistent memory

Google senior AI product manager Shubham Saboo has turned one of the thorniest problems in agent design into an open-source engineering exercise: persistent memory. This week, he published an ...
Hosted on MSN

Copilot Wave 3 and ChatGPT Memory Overhaul Transform Your Daily Workflow

As of May 2026, the AI assistant landscape has evolved beyond simple chatbots into sophisticated workflow engines. Microsoft's Copilot Wave 3 introduces a cross-vendor multi-model architecture, while ...
Forbes

AI Agent Types – And Memory

Forbes contributors publish independent expert analyses and insights. I am an MIT Senior Fellow & Lecturer, 5x-founder & VC investing in AI In the big conversation that companies and people are having ...
Hosted on MSN

AI Agents Are Rewriting Your Daily Workflow in 2026

If you’re still treating AI assistants as glorified search engines or text generators, you’re missing the point of 2026. This year marked a decisive shift: ChatGPT rolled out persistent memory for ...
9to5Mac

Anthropic updates Claude Managed Agents with three new features

The first new feature is called dreaming, which Anthropic classifies as a research preview. Anthropic says dreaming extends Claude’s memory capabilities “by reviewing past sessions to find patterns ...